1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
|
# Copyright (C) 2006 Canonical Ltd
#
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation; either version 2 of the License, or
# (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program; if not, write to the Free Software
# Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
import BaseHTTPServer
import errno
import os
from SimpleHTTPServer import SimpleHTTPRequestHandler
import socket
import posixpath
import random
import re
import sys
import threading
import time
import urllib
import urlparse
from bzrlib.transport import Server
from bzrlib.transport.local import LocalURLServer
class WebserverNotAvailable(Exception):
pass
class BadWebserverPath(ValueError):
def __str__(self):
return 'path %s is not in %s' % self.args
class TestingHTTPRequestHandler(SimpleHTTPRequestHandler):
"""Handles one request.
A TestingHTTPRequestHandler is instantiated for every request
received by the associated server.
"""
def log_message(self, format, *args):
tcs = self.server.test_case_server
tcs.log('webserver - %s - - [%s] %s "%s" "%s"',
self.address_string(),
self.log_date_time_string(),
format % args,
self.headers.get('referer', '-'),
self.headers.get('user-agent', '-'))
def handle_one_request(self):
"""Handle a single HTTP request.
We catch all socket errors occurring when the client close the
connection early to avoid polluting the test results.
"""
try:
SimpleHTTPRequestHandler.handle_one_request(self)
except socket.error, e:
if (len(e.args) > 0
and e.args[0] in (errno.EPIPE, errno.ECONNRESET,
errno.ECONNABORTED,)):
self.close_connection = 1
pass
else:
raise
_range_regexp = re.compile(r'^(?P<start>\d+)-(?P<end>\d+)$')
_tail_regexp = re.compile(r'^-(?P<tail>\d+)$')
def parse_ranges(self, ranges_header):
"""Parse the range header value and returns ranges and tail.
RFC2616 14.35 says that syntactically invalid range
specifiers MUST be ignored. In that case, we return 0 for
tail and [] for ranges.
"""
tail = 0
ranges = []
if not ranges_header.startswith('bytes='):
# Syntactically invalid header
return 0, []
ranges_header = ranges_header[len('bytes='):]
for range_str in ranges_header.split(','):
# FIXME: RFC2616 says end is optional and default to file_size
range_match = self._range_regexp.match(range_str)
if range_match is not None:
start = int(range_match.group('start'))
end = int(range_match.group('end'))
if start > end:
# Syntactically invalid range
return 0, []
ranges.append((start, end))
else:
tail_match = self._tail_regexp.match(range_str)
if tail_match is not None:
tail = int(tail_match.group('tail'))
else:
# Syntactically invalid range
return 0, []
return tail, ranges
def send_range_content(self, file, start, length):
file.seek(start)
self.wfile.write(file.read(length))
def get_single_range(self, file, file_size, start, end):
self.send_response(206)
length = end - start + 1
self.send_header('Accept-Ranges', 'bytes')
self.send_header("Content-Length", "%d" % length)
self.send_header("Content-Type", 'application/octet-stream')
self.send_header("Content-Range", "bytes %d-%d/%d" % (start,
end,
file_size))
self.end_headers()
self.send_range_content(file, start, length)
def get_multiple_ranges(self, file, file_size, ranges):
self.send_response(206)
self.send_header('Accept-Ranges', 'bytes')
boundary = "%d" % random.randint(0,0x7FFFFFFF)
self.send_header("Content-Type",
"multipart/byteranges; boundary=%s" % boundary)
self.end_headers()
for (start, end) in ranges:
self.wfile.write("--%s\r\n" % boundary)
self.send_header("Content-type", 'application/octet-stream')
self.send_header("Content-Range", "bytes %d-%d/%d" % (start,
end,
file_size))
self.end_headers()
self.send_range_content(file, start, end - start + 1)
self.wfile.write("--%s\r\n" % boundary)
def do_GET(self):
"""Serve a GET request.
Handles the Range header.
"""
path = self.translate_path(self.path)
ranges_header_value = self.headers.get('Range')
if ranges_header_value is None or os.path.isdir(path):
# Let the mother class handle most cases
return SimpleHTTPRequestHandler.do_GET(self)
try:
# Always read in binary mode. Opening files in text
# mode may cause newline translations, making the
# actual size of the content transmitted *less* than
# the content-length!
file = open(path, 'rb')
except IOError:
self.send_error(404, "File not found")
return
file_size = os.fstat(file.fileno())[6]
tail, ranges = self.parse_ranges(ranges_header_value)
# Normalize tail into ranges
if tail != 0:
ranges.append((file_size - tail, file_size))
self._satisfiable_ranges = True
if len(ranges) == 0:
self._satisfiable_ranges = False
else:
def check_range(range_specifier):
start, end = range_specifier
# RFC2616 14.35, ranges are invalid if start >= file_size
if start >= file_size:
self._satisfiable_ranges = False # Side-effect !
return 0, 0
# RFC2616 14.35, end values should be truncated
# to file_size -1 if they exceed it
end = min(end, file_size - 1)
return start, end
ranges = map(check_range, ranges)
if not self._satisfiable_ranges:
# RFC2616 14.16 and 14.35 says that when a server
# encounters unsatisfiable range specifiers, it
# SHOULD return a 416.
file.close()
# FIXME: We SHOULD send a Content-Range header too,
# but the implementation of send_error does not
# allows that. So far.
self.send_error(416, "Requested range not satisfiable")
return
if len(ranges) == 1:
(start, end) = ranges[0]
self.get_single_range(file, file_size, start, end)
else:
self.get_multiple_ranges(file, file_size, ranges)
file.close()
def translate_path(self, path):
"""Translate a /-separated PATH to the local filename syntax.
If the server requires it, proxy the path before the usual translation
"""
if self.server.test_case_server.proxy_requests:
# We need to act as a proxy and accept absolute urls,
# which SimpleHTTPRequestHandler (parent) is not
# ready for. So we just drop the protocol://host:port
# part in front of the request-url (because we know
# we would not forward the request to *another*
# proxy).
# So we do what SimpleHTTPRequestHandler.translate_path
# do beginning with python 2.4.3: abandon query
# parameters, scheme, host port, etc (which ensure we
# provide the right behaviour on all python versions).
path = urlparse.urlparse(path)[2]
# And now, we can apply *our* trick to proxy files
path += '-proxied'
return self._translate_path(path)
def _translate_path(self, path):
return SimpleHTTPRequestHandler.translate_path(self, path)
if sys.platform == 'win32':
# On win32 you cannot access non-ascii filenames without
# decoding them into unicode first.
# However, under Linux, you can access bytestream paths
# without any problems. If this function was always active
# it would probably break tests when LANG=C was set
def _translate_path(self, path):
"""Translate a /-separated PATH to the local filename syntax.
For bzr, all url paths are considered to be utf8 paths.
On Linux, you can access these paths directly over the bytestream
request, but on win32, you must decode them, and access them
as Unicode files.
"""
# abandon query parameters
path = urlparse.urlparse(path)[2]
path = posixpath.normpath(urllib.unquote(path))
path = path.decode('utf-8')
words = path.split('/')
words = filter(None, words)
path = os.getcwdu()
for word in words:
drive, word = os.path.splitdrive(word)
head, word = os.path.split(word)
if word in (os.curdir, os.pardir): continue
path = os.path.join(path, word)
return path
class TestingHTTPServer(BaseHTTPServer.HTTPServer):
def __init__(self, server_address, RequestHandlerClass,
test_case_server):
BaseHTTPServer.HTTPServer.__init__(self, server_address,
RequestHandlerClass)
# test_case_server can be used to communicate between the
# tests and the server (or the request handler and the
# server), allowing dynamic behaviors to be defined from
# the tests cases.
self.test_case_server = test_case_server
def server_close(self):
"""Called to clean-up the server.
Since the server may be in a blocking read, we shutdown the socket
before closing it.
"""
self.socket.shutdown(socket.SHUT_RDWR)
BaseHTTPServer.HTTPServer.server_close(self)
class HttpServer(Server):
"""A test server for http transports.
Subclasses can provide a specific request handler.
"""
# Whether or not we proxy the requests (see
# TestingHTTPRequestHandler.translate_path).
proxy_requests = False
# used to form the url that connects to this server
_url_protocol = 'http'
# Subclasses can provide a specific request handler
def __init__(self, request_handler=TestingHTTPRequestHandler):
Server.__init__(self)
self.request_handler = request_handler
self.host = 'localhost'
self.port = 0
self._httpd = None
def _get_httpd(self):
if self._httpd is None:
self._httpd = TestingHTTPServer((self.host, self.port),
self.request_handler,
self)
host, self.port = self._httpd.socket.getsockname()
return self._httpd
def _http_start(self):
httpd = self._get_httpd()
self._http_base_url = '%s://%s:%s/' % (self._url_protocol,
self.host,
self.port)
self._http_starting.release()
while self._http_running:
try:
httpd.handle_request()
except socket.timeout:
pass
def _get_remote_url(self, path):
path_parts = path.split(os.path.sep)
if os.path.isabs(path):
if path_parts[:len(self._local_path_parts)] != \
self._local_path_parts:
raise BadWebserverPath(path, self.test_dir)
remote_path = '/'.join(path_parts[len(self._local_path_parts):])
else:
remote_path = '/'.join(path_parts)
return self._http_base_url + remote_path
def log(self, format, *args):
"""Capture Server log output."""
self.logs.append(format % args)
def setUp(self, backing_transport_server=None):
"""See bzrlib.transport.Server.setUp.
:param backing_transport_server: The transport that requests over this
protocol should be forwarded to. Note that this is currently not
supported for HTTP.
"""
# XXX: TODO: make the server back onto vfs_server rather than local
# disk.
assert backing_transport_server is None or \
isinstance(backing_transport_server, LocalURLServer), \
"HTTPServer currently assumes local transport, got %s" % \
backing_transport_server
self._home_dir = os.getcwdu()
self._local_path_parts = self._home_dir.split(os.path.sep)
self._http_starting = threading.Lock()
self._http_starting.acquire()
self._http_running = True
self._http_base_url = None
self._http_thread = threading.Thread(target=self._http_start)
self._http_thread.setDaemon(True)
self._http_thread.start()
# Wait for the server thread to start (i.e release the lock)
self._http_starting.acquire()
self._http_starting.release()
self.logs = []
def tearDown(self):
"""See bzrlib.transport.Server.tearDown."""
self._httpd.server_close()
self._http_running = False
self._http_thread.join()
def get_url(self):
"""See bzrlib.transport.Server.get_url."""
return self._get_remote_url(self._home_dir)
def get_bogus_url(self):
"""See bzrlib.transport.Server.get_bogus_url."""
# this is chosen to try to prevent trouble with proxies, weird dns,
# etc
return 'http://127.0.0.1:1/'
class HttpServer_urllib(HttpServer):
"""Subclass of HttpServer that gives http+urllib urls.
This is for use in testing: connections to this server will always go
through urllib where possible.
"""
# urls returned by this server should require the urllib client impl
_url_protocol = 'http+urllib'
class HttpServer_PyCurl(HttpServer):
"""Subclass of HttpServer that gives http+pycurl urls.
This is for use in testing: connections to this server will always go
through pycurl where possible.
"""
# We don't care about checking the pycurl availability as
# this server will be required only when pycurl is present
# urls returned by this server should require the pycurl client impl
_url_protocol = 'http+pycurl'
|